• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

DPAD Daily user stats 14-1-09

Assimilator1

Assimilator1

Elite Member
Output by local teams (Mpts) (No other teams are currently close to us in total points and output! 🙂).

TA - 8,186,056 😎 (133 million ahead)
DPC - 873,697

Prod Rank __ Team Rank __ Days Mpts _________ ______ Mpts ___________ Name
_1 _________ _1 _________ 7,199,781 _________ 1,183,622,131 _________ amd.borg
_2 _________ _2 _________ _ 726,252 _________ _ 358,125,716 _________ JonB
_3 _________ _4 _________ __ 76,758 _________ _ 124,354,893 _________ Amaroque
_4 _________ _5 _________ __ 75,316 _________ _ 113,402,837 _________ Silverthorne
_5 _________ 46 _________ __ 59,367 _________ ___ 3,659,884 _________ Chocobo Navy
_6 _________ 13 _________ __ 48,582 _________ __ 23,627,906 _________ caferace

Active members over the past week - 12 (-1 from last stats)

[TA]amd.borg
[TA]JonB
[TA]Amaroque
[TA]Silverthorne
[TA]viztech
[TA]compudog
[TA]Assimilator1
[TA]OrangeKid
[TA]Petrusbroder
[TA]caferace
[TA]Pacemaker
[TA]Chocobo Navy

DPAD stats collected from Free-DC ('yesterday' column).
Official TeAm stats from the DPAD site.

*****************************************************************************************

Another awesome dump by Borgie!:Q😀

I've always wondered how much difference RAM speed, RAM timings & chipset timings would make to DPAD so I've been tweaking my quad rig a bit lately 🙂. I thought a few of you might find it interesting & I also know many here don't go to the DPAD forums so I'll post my results so far as well.

My Q6600 runs on an Asus P5Q Pro mbrd (Intel P45 chipset), until a few days ago it was running at 3.35 GHz, 372 MHz FSB, 745MHz RAM (OCZ DDR2 800 CL4) with 4-4-4-15 timings & tRD 11.
That setup scored an average of 1160.6 Kpts/s (measured using muon1benchmark).

Now I wasn't happy with the RAM being underclocked (due to divider choises & previously no time to tweak), so the 1st tweak I did was to the RAM speed, I bumped it upto 933 MHz with 5-5-5-15 timings, tRD was also automatically reduced to 10. The CAS Latency time for this was 10.75ns, which coincidently was the same as the previous settings so no extra CAS latency penalty 🙂. vRAM had to be boosted from ~1.95v to 2.09v to be stable at this speed.This scored an average of 1213 Kpts/s, which is ~4.5% faster, not a huge improvement but still a worthwhile one 🙂.

After reading an article in a review by Anandtech regarding tRD & it's apparent big gains in memory performance improvement I thought I'd give that a go too.So the next setup was as above except I cut tRD from 10 down to 7.
This scored an average of 1267 Kpts/s, ironically another ~4.5% speed gain 🙂.

My next step is to try a higher FSB with a lower multiplier, so my rig now runs at 3.34 GHz, 417 MHz FSB, 834 MHz RAM with 4-4-4-15 timings & TRD 8 (tRD 7 might be possible but I'll try that latter). However I only just finished stabilty testing that about 1hr ago so I don't have results yet. I'll give an update tomorrow with the new figures.
Hopefully a combination of a higher FSB, 1:1 RAM/FSB ratio & lower CAS time of 9.58ns will give quite an improvement (although the lower(?) memory bandwidth may counter that a bit), time will tell!........

[update] Well this sucks!, despite much higher FSB & lower RAM latencies it only scores 1258.4 Kpts/s, worse than the 372:933 setting. Oh well on the plus side dropping the FSB down means I can set the NBv back to def.

Although I now have some DDR2 1066 RAM to play with 😀, I wonder which FSB will be quicker then? maybe the one with the highest RAM speed??

Best I could get my RAM stable at on a 372 MHz FSB was 991 MHz DDR, 5-5-5-15 timings, tRD7. Average score over a 21.2 hr run was 1278.1 Kpts/s, slightly higher at ~1% faster than the previous best 😛.[update] Retested & I got 1256.4 Kpts/s, wth??😕

Currently stability testing 417 MHz FSB, Q6600 @3.34 GHz, 1046 MHz RAM, 5-5-5-18, tRD 8.
Eventually got it to pass OCCT for 24hrs+.
Bizarely this setup only scored 1273.3 Kpt/s ..... that doesn't make any sense 😕 (score updated from more results, pushed score up 8).[update] makes sense now the other score has dropped!
 
I thinks I'm gonna need to play with settings with that benchmark tool - I need to replace my heatsink or do a reapplication of TIM before I put my new chip to crunching at 4.5ghz at 560fsb
 
Originally posted by: GLeeM
Thanks for the stats, Assimilator1 🙂

Nice rundown on your ram tweaks!
Click to expand...

Interesting stuff AS1!:beer:

I've got a Gigabyte MB (GA-EP35C-DS3R) that will take either DDR2 or DDR3 RAM (both dual channel). I'm wondering if it would be worth tossing in a couple of DDR3 sticks to get an increase in performance.

cheers and nice job TA DPAD crunchers!:beer:
 
nefariouscaine
Your chip can do 4.5 GHz?!:shocked:😎, wow that's one awesome o/c!😀, at that speed your CPU would be the number 1 dualie in the DPAD chart by miles!
Err although your sig says 4.2GHz 560FSB?....

biodoc
An Intel mbrd that can take either RAM? cool! I knew it was technically possible but I hadn't heard of a mbrd that had actually implemented it, not that I read hoards of mbrd reviews though 😉.
I was just wondering about the price of DDR3 RAM, I knew it wasn't cheap so I did a quick lookup of prices here in the UK vs DDR2.

2x2GB OCZ/Kingston/Crucial DDR2 800 CL5 £33-35
2x2GB OCZ DDR2 800 CL4 £45
2x2GB OCZ DDR2 1066 CL5 £55
2x2GB OCZ DDR3 1333 CL9 £92
2x2GB Corsair DDR3 1600 CL9 £142 :Q
2x2GB OCZ DDR3 1600 CL8 £166 :shocked:

I don't know that any of the DDR3 would be worth it (ones listed here anyway), the 1333 just has too high a latency, 13.5ns vs 9.38ns of the 1066 DDR2. The 1600 would probably perform better but what a price!:Q (btw it's CAS latency time is 10ns).
Maybe the prices in the USA are better??

Btw what spec RAM have you got atm?
 
I cheaped out when I bought my DDR2 800 RAM:

# Cas Latency: 5
# Timing: 5-6-6-18

Yeah, it's an interesting MB: it has 4 DDR2 slots (8GB max) and 2 DDR3 slots (max 4GB).

DDR3 support for 1333/1066/800 MHz.

Newegg prices in the states:

For 2 x 2GB DDR3 1066 (range $100-$150)

# Cas Latency: 7
# Timing: 7-7-7-21
# Voltage: 1.5V

For 2 x 2GB DDR3 1333 ($125 AR) link

# Cas Latency: 6
# Timing: 6-5-5-18
# Voltage: 1.85V

Bottom line: Still quite expensive but prices are dropping!

EDIT: Since my MB does not support DDR3 1600 then my max OC for a Q6600 is only 3 GHz!🙁

No use thinking about it anymore!:laugh:
 
I was going to say be wary of buying high latency DDR3, could end up being slower!
Formula for calculating CAS latency in ns is (CLx2000)/RAM speed(DDR MHz).

And yea because of your DDR3 limitations you'll get a better CPU o/c on DDR2.
I've found that OCZ Platinum DDR2 1066 CL 5-5-5-18 @2.2v is relatively cheap & high peforming RAM 😉.
 
Update on my RAM settings.

Tried running my new 1066 RAM at 1191 MHz (372 MHz FSB) but I couldn't even get it to boot!, so then tried for 1119 MHz but I couldn't get it stable, which is dissapointing 🙁. Although I haven't ran out of options yet to get it stable.
 
Well I don't know wth's going here with DPAD benchmark, I was suspucious that I was getting varying results even at the same settings so I retested with 372:495, this time I got 1256.4 Kpts/s 😕
Maybe it's just not accurate enough to measure small differences 🙁

Guess I'd better do some new stats! 😉
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top